ARTIFICIAL INTELLIGENCE: A
COMPREHENSIVE OVERVIEW
Aishwarya Anand Arora
(Assistant Professor)
Dept. of Computer Science
School of Open Learning, University of Delhi
Source: https://www.pangaeax.com/2022/05/09/ai-types-and-applications-in-business/
INTRODUCTION TO AI
AI\ML\DL
WHAT IS ARTIFICIAL INTELLIGENCE?
• Human-like (“How to simulate humans’ intellect and behavior on by a
machine.)
• Mathematical problems (puzzles, games, theorems)
• Common-sense reasoning (if there is parking-space, probably illegal to
park)
• Expert knowledge: lawyers, medicine, diagnosis
• Social behavior
• Rational-like:
• achieve goals, have performance measure
WHAT IS ARTIFICIAL INTELLIGENCE
• Thought processes
• “The exciting new effort to make computers think . Machines
with minds, in the full and literal sense” (Haugeland, 1985)
• Behavior
• “The study of how to make computers do things at which, at
the moment, people are better.” (Rich, and Knight, 1991)
COMPONENTS OF AI
Problem-
Learning Reasoning
solving
Language
Perception understandin
g.
HISTORY OF AI
RISE AGAIN
HOW AI WORKS
AI IN NLP
• Language models,
• Sentiment analysis
• Machine translation.
• OpenAI’s GPT series.
• Google’s BERT model.
AI IN COMPUTER
VISION
• Image and video analysis
• Facial recognition
• Autonomous navigation.
• AI in autonomous vehicles.
RECENT
TRENDS
AREAS OF APPLICATION
AI IN HEALTHCARE
Applications: Diagnostics, personalized treatment, patient care.
Examples:
• IBM Watson Health for cancer treatment.
• AI in radiology for image analysis.
AI IN FINANCE
Applications: Fraud detection, algorithmic trading, personalized banking.
Examples:
• JPMorgan’s COiN for contract review.
• Robo-advisors for investment management.
AI IN TRANSPORTATION
Applications: Autonomous vehicles, traffic management, predictive
maintenance.
Examples:
• Tesla Autopilot.
• Waymo’s self-driving cars.
AI IN RETAIL
Applications: Personalized shopping experiences, inventory management,
supply chain optimization.
Examples:
• Amazon’s recommendation engine.
• AI-powered cashier-less stores.
AI IN MANUFACTURING
Applications: Predictive maintenance, quality control, robotics.
Examples:
• Siemens’ AI-driven automation.
• GE’s use of AI for predictive maintenance.
AI IN ENTERTAINMENT
Applications: Content recommendation, creation, and curation.
Examples:
• Netflix’s recommendation algorithm.
• AIgenerated music and art.
AI IN EDUCATION
Applications: Personalized learning, administrative tasks, virtual tutoring.
Examples:
• Intelligent tutoring systems.
• Automated grading systems.
AI IN AGRICULTURE
Applications: Precision farming, crop monitoring, pest control.
Examples:
• John Deere’s AI-driven equipment.
• AIpowered drones for crop monitoring.
AI IN SECURITY
Applications: hreat detection, cybersecurity, surveillance.
Examples:
• Darktrace for cybersecurity.
• AI in surveillance systems.
AI IN CUSTOMER SERVICE
Applications: Chatbots, virtual assistants, automated support.
Examples:
• Amazon Alexa.
• Google Assistant.
AI LIMITATIONS
• Data Dependency: AI systems require large amounts of high-quality data to function
effectively.
• Generalization: They are typically good at specific tasks but lack the ability to transfer
knowledge across different domains.
• Explainability: AI operate as "black boxes," making it difficult to understand how they
arrive at specific decisions or predictions.
• Robustness: AI systems can be sensitive to small changes in input data, leading to
unexpected or incorrect outputs.
• Integration Challenges: Integrating AI into existing systems and workflows can be
challenging, requiring significant changes in infrastructure and processes.
https://data-flair.training/blogs/wp-content/uploads/sites/2/2019/09/Pros-and-Cons-of-Artificial-Intelligence.jpg
AI ETHICS
Fairness Transparenc
y
Accountabilit
Privacy
y
Beneficence Autonomy
AI BIAS
• Training Data Bias: If the data used to train an AI system contains biases, the
system is likely to replicate those biases.
• Algorithmic Bias: Even if the training data is unbiased, the algorithms
themselves might introduce bias through the way they process information.
• Deployment Bias: Bias can also be introduced during the deployment of AI
systems,
CAN AI REPLACE HUMAN
AI FUTURE
• Develop techniques to identify, measure, and eliminate biases in AI models
and training data.
• Create models that provide clear, understandable explanations for their
decisions.
• Enhance techniques that allow models to apply knowledge from one task to
new, related tasks.
• Develop methods to make AI models more resilient to adversarial attacks, data
noise, and unexpected inputs.
• Establish frameworks, policies, and standards for the ethical use of AI.
WHAT IS YOUR VIEW?
